Sang-Kuk Lee
From MIPAL
Sang-Kuk Lee (이상국)
|
Research Interests
- Machine Learning, Computer Vision
- Multi-modal Deep Learning in Video
- Socially-useful technology
Education
- B.S. Department of Nuclear Engineering, Seoul National University, Korea, 2002-2008
Work Experiences
- R&D of target detection and tracking algorithm in the Radar system, Hope Air Tec Co.Ltd., 2009-2013
- Object recognition & Network system programmer in TeamSNU on the DARPA Robotics Challenge(DRC), 2014-2015
- Founder & CEO in eyear , 2016-2017
- Cofounder in LetinAR Inc.
- Founder & CEO in V.DO Inc., 2017.6 ~
Publications
- Sangkuk Lee, Sung Joon Son, Jiyong Oh and Nojun Kwak, " Handwritten Music Symbol Classification Using Deep Convolutional Neural Networks", Proc. The 3rd International Conference on Information Science and Security (ICISS2016), Accepted, 2016.
- Jiyong Oh, Sung Joon Son, Sangkuk Lee, Ji-Won Kwon and Nojun Kwak, "Online recognition of handwritten music symbols", International Journal of Document Analysis and Recognition(IJDAR), accepted for publication, 2017
- Sangkuk Lee, Daesik Kim, Myunggi Lee, Jihye Hwang and Nojun Kwak, "Where to Play: Retrieval of video segments using natural-language queries", arXiv, July 2017.
- Daesik Kim, Youngjoon Yoo, Jeesoo Kim, Sangkuk Lee and Nojun Kwak, "Dynamic Graph Generation Network: Generating Relational Knowledge from Diagrams", arXiv, Nov. 2017.
Competition
- 4th place of "Textbook Question Answering Challenge", CVPR 2017 Workshop on Visual Understanding Across Modalities